| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-02-27 | |||
| 17:04:47 | mriedem | in the project def too | |
| 17:05:10 | mriedem | the nnet job should be explicit about using nova-network since d-g defaults to using neutron if not specified since ocata | |
| 17:06:12 | sdague | mriedem: there is no explicit nnet | |
| 17:06:27 | mriedem | the nnet job is explicitly nova-network right? | |
| 17:06:33 | mriedem | hence the rename | |
| 17:06:44 | mriedem | you should set "export DEVSTACK_GATE_NEUTRON=0" in those jobs | |
| 17:07:16 | mriedem | because of https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate-wrap.sh#L206 | |
| 17:07:37 | sdague | ah, I didn't realize that happened | |
| 17:07:48 | mriedem | yeah, part of getting nova-net out of ocata jobs | |
| 17:08:00 | mriedem | clarkb and i tag teamed like the hart brothers | |
| 17:10:29 | sdague | how about now? | |
| 17:10:52 | mriedem | you missed one | |
| 17:12:31 | sdague | gah, unsaved buffer | |
| 17:16:59 | openstackgerrit | JeffreyGuan proposed openstack/python-novaclient master: Grammar typo in the comments for function set_meta. https://review.openstack.org/438631 | |
| 17:18:26 | mriedem | dansmith: backports are up https://review.openstack.org/#/q/Ic96a5eb3728f97a3c35d2c5121e6fdcd4fd1c70b,n,z | |
| 17:21:47 | mriedem | sdague: +1 now | |
| 17:25:57 | openstackgerrit | John L. Villalovos proposed openstack/nova master: Use flake8-import-order https://review.openstack.org/435575 | |
| 17:30:40 | dansmith | mriedem: awesome thanks | |
| 17:39:51 | dansmith | mriedem: he found another couple things, but they're newton and beyond so we should be good on mitaka once those merge | |
| 17:40:07 | mriedem | word to your mother | |
| 17:40:20 | mriedem | reminds me i was going to write a spec about exposing the service version in the REST API | |
| 17:42:41 | openstackgerrit | John Garbutt proposed openstack/nova-specs master: WIP: Backlog spec on preemptible instances https://review.openstack.org/438640 | |
| 17:57:01 | dansmith | mriedem: actually I think we already fixed the other things he hit in newton, but I'll confirm with him tomorrow when he's back | |
| 17:59:56 | cfriesen | mriedem: here's the stable/newton release note for bug 1644248 | |
| 17:59:56 | openstack | bug 1644248 in OpenStack Compute (nova) ocata "Nova incorrectly tracks live migration progress" [High,In progress] https://launchpad.net/bugs/1644248 - Assigned to Matt Riedemann (mriedem) | |
| 18:00:11 | cfriesen | mriedem: https://review.openstack.org/438650 | |
| 18:00:23 | mriedem | ok | |
| 18:09:55 | openstackgerrit | Maciej Szankin proposed openstack/nova master: Remove invalid tests-py3 whitelist item https://review.openstack.org/438660 | |
| 18:11:15 | openstackgerrit | Maciej Szankin proposed openstack/nova master: Remove invalid tests-py3 whitelist item https://review.openstack.org/438660 | |
| 18:12:43 | macsz | johnthetubaguy: https://review.openstack.org/#/c/438660/ | |
| 18:14:06 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/438309 | |
| 18:18:08 | sfinucan | mriedem: RE: the cliff vs. click thing - sorry I missed that discussion (we had gifts to get) | |
| 18:18:10 | openstackgerrit | Maciej Szankin proposed openstack/nova master: Tests: remove .testrepository/times.dbm in tox.ini https://review.openstack.org/435111 | |
| 18:18:14 | sfinucan | but I've comments left on https://review.openstack.org/#/c/433603/2 | |
| 18:18:55 | sfinucan | tl;dr: IMO, cliff is broken by design and no one should be using it. In addition, click offers enough cool things to justify its inclusion in g-r | |
| 18:20:00 | sfinucan | I'd be happy to try talk the openstackclient team around, but without "code on the table" I doubt I'd get far :) | |
| 18:20:35 | openstackgerrit | Maciej Szankin proposed openstack/nova master: conf: remove *_topic config opts https://review.openstack.org/433924 | |
| 18:21:26 | sdague | sfinucan: well, feel free to have the conversation with the openstack client team, but definitely personally -1 on replacing one different CLI system with another one, instead of standardizing into what everyone else in the community is using | |
| 18:21:57 | dansmith | me too | |
| 18:22:35 | openstackgerrit | John L. Villalovos proposed openstack/nova master: Use flake8-import-order https://review.openstack.org/435575 | |
| 18:24:01 | sfinucan | sdague: I do think they're standardizing on a bad choice but that's understandable | |
| 18:25:25 | sdague | sfinucan: well, there is a ton of experience in the community, cliff was adopted into openstack as an openstack project, so am very much not interested in running off in a different direction | |
| 18:27:20 | sfinucan | sdague: Yup, standardizing on things is good. It's just that I think cliff is a poor choice to standardize on | |
| 18:29:00 | sdague | sfinucan: ok, well get in a time machine and go back to 2013 and have that argument :) | |
| 18:29:20 | sfinucan | :) | |
| 18:51:47 | openstackgerrit | Stephen Finucane proposed openstack/nova-specs master: Reserve NUMA nodes with PCI devices attached https://review.openstack.org/364468 | |
| 18:55:19 | openstackgerrit | Chris Friesen proposed openstack/python-novaclient master: Add support for showing flavor details (v2.43) https://review.openstack.org/435141 | |
| 19:02:14 | mriedem | cfriesen: i added the co-authored-by on your newton reno | |
| 19:02:21 | mriedem | +2 now | |
| 19:08:25 | cfriesen | mriedem: oops, missed that. thanks for the tweak | |
| 19:33:01 | openstackgerrit | Sujitha proposed openstack/nova master: Remove mox from nova.tests.unit.virt.xenapi.test_vmops.GetConsoleOutputTestCase https://review.openstack.org/438678 | |
| 19:35:04 | openstackgerrit | Sujitha proposed openstack/nova master: Remove mox from nova.tests.unit.virt.xenapi.test_vmops[1] https://review.openstack.org/438678 | |
| 20:21:15 | mriedem | sdague: did you ever find a patch to remove LocalFS? | |
| 20:21:26 | mriedem | i couldn't find one of yours like that | |
| 20:21:39 | sdague | I might have abanoned it, let me go look | |
| 20:22:26 | sdague | https://review.openstack.org/#/c/324720/3 - just restored and recheck | |
| 20:46:25 | mriedem | melwitt: on https://review.openstack.org/#/c/431749/ - was there anything you wanted to add to that given the discussion at the PTG about adding project/user IDs to the placement API for handling quota when a cell is down? | |
| 21:04:31 | mriedem | sdague: just checking, but did you agree to take a stab at a spec for storing limits in keystone? | |
| 21:05:11 | sdague | mriedem: yeh, I need to go look at the keystone spec | |
| 21:05:25 | mriedem | ok, just updating some notes | |
| 21:05:25 | mriedem | thanks | |
| 21:07:04 | cfriesen | what's the current view on cleaning up stale data in the nova db? is there a way to clean up all the tables, or just some? And is anyone working on improving it? | |
| 21:07:31 | mriedem | cfriesen: there is the archive command | |
| 21:07:35 | mriedem | but that just moves to shadow tables | |
| 21:07:46 | mriedem | there was a spec/effort at one point for adding a purge instances command | |
| 21:07:53 | mriedem | but withered on the vine, needs an owner | |
| 21:09:02 | cfriesen | mriedem: this one? https://blueprints.launchpad.net/nova/+spec/purge-deleted-instances-cmd | |
| 21:09:48 | mriedem | yeah | |
| 21:09:58 | mriedem | spec would have to be re-proposed for pike | |
| 21:27:43 | mriedem | raj_singh: sdague: efried: i'm seeing several rechecks on https://review.openstack.org/#/c/412634/ but the py35 dsvm job failures look legit | |
| 21:27:49 | mriedem | http://logs.openstack.org/34/412634/25/check/gate-tempest-dsvm-py35-ubuntu-xenial/c7c593e/logs/screen-n-api.txt.gz?level=TRACE | |
| 21:29:08 | openstackgerrit | Sean Dague proposed openstack/nova master: Adopts keystoneauth with glance client. https://review.openstack.org/412634 | |
| 21:29:29 | sdague | mriedem: ok, let me rebase to master to see if it was keeping a very odd old parent on some things | |
| 21:30:56 | mriedem | ok, the failure isn't clear to me otherwise | |
| 21:32:26 | mriedem | http://logs.openstack.org/34/412634/25/check/gate-tempest-dsvm-py35-ubuntu-xenial/c7c593e/logs/screen-g-api.txt.gz#_2017-02-24_03_57_52_828 is the g-api response i think | |
| 21:32:27 | sdague | mriedem: yeh, the patch being from Feb 13 means that I expect there remained broken things on the python 35 side in nova | |
| 21:32:32 | sdague | that only fixed much later | |
| 21:32:57 | sdague | mriedem: ok, well lets see what a clean rebase gives us | |
| 21:33:19 | sdague | I'd rather not chase until we know we're working off a base that we know works in python3.5 | |
| 21:33:36 | mriedem | sure | |
| 21:33:44 | mriedem | also, we have a way to enable glanceclient debug if we need to debug this | |
| 21:33:53 | mriedem | since it's not tracing the requests/responses | |
| 21:34:13 | mriedem | https://github.com/openstack/nova/blob/master/nova/conf/glance.py#L87 | |
| 21:35:00 | mriedem | dansmith: i can drop this now right? https://review.openstack.org/#/c/420976/ | |
| 21:35:03 | mriedem | you're using something else | |
| 21:35:11 | dansmith | mriedem: yeah | |
| 21:35:32 | raj_singh | sdague: thx. If this rebase does not fix then I will debug it tomorrow. | |
| 21:35:38 | mriedem | dansmith: and this? https://review.openstack.org/#/c/421399/ | |
| 21:36:19 | dansmith | mriedem: no, I'll have to have that when I make this spawn conductors on the subnode | |
| 21:36:35 | mriedem | ok | |
| 21:36:47 | sdague | raj_singh: sounds great | |
| 21:57:43 | mriedem | dansmith: L22 in https://etherpad.openstack.org/p/nova-ptg-pike-cells - can you link that in there? | |
| 21:58:02 | dansmith | jess | |
| 21:58:07 | mriedem | i thought that was https://review.openstack.org/#/c/436094/ | |
| 21:58:11 | mriedem | but wasn't sure | |
| 21:58:30 | dansmith | yup | |
| 22:14:36 | mriedem | the super-conductor / fleet stuff is really not discussed at all in https://etherpad.openstack.org/p/nova-ptg-pike-cells | |
| 22:14:56 | mriedem | so i'm at a loss for notes besides wanting to move all conductors in the multinode job to the subnode | |
| 22:15:14 | mriedem | i don't have the whiteboard picture either | |
| 22:15:20 | mriedem | did that get posted anywhere? | |